The purpose of this research is to develop procedures for assessing family interaction related to disordered child behavior (of a mild to moderate degree of severity), and to develop procedures for efficiently modifying such interaction and maintaining these changes. Two studies are underway: 1) Families with recurring mother-child problem-interactions are assigned randomly to one of three conditions: a) father included in the intervention, b) father not included, and c) minimum treatment wait-control. Intervention consists of 4 - 8 sessions of training in conflict resolution taught by modeling and behavior rehearsal, and training in contingency management. Pre-, post- and six-month follow-up assessments are conducted. 2) This study is designed to answer the question - with what kinds of parent-child problems does conflict resolution training, or contingency management, or some combination of both work best?